|
|
Robotic leg inverse dynamic modelling and design optimization in Simmechanics
Grepl, Robert ; Kratochvíl, Ctirad
This paper deals with the design of inverse dynamical model of pantographic robotic leg in Matlab - SimMechanics. This problem belongs to the class of dynamical inverse modelling, where the positions ofactuators and efector are defined in different coordinate systems. Therefore we also have to use inverse kinematicsmodel. This inverse dynamic model serves as the base of robotic leg design optimization. The definition of valuefunction is based on known properties of drives and variable parameters of the leg design. Optimization maximizesparameters of efector trajectory.
|
|
| |
|
| |
|
| |
|
| |
|
|
Mechatronics robotics and biomechanics 2001 /3./. Proceedings
Kotek, Vladimír ; Kratochvíl, Ctirad ; Ehrenberger, Zdeněk
The conference covers main area of mechatronics, robotics and biomechanics, i.e.: Philosophical aspects of Mechatronics, Robotics and Biomechanics, Application of arificial intelligence in these fields, Modelling and control of these systens, Smart sytems, Education problems of Mechatronics, Biomechanics of skeleton - muscular system, Cardio - vascular and hearing system biomechanics.
|
|
| |
|
| |
|
| |
|
|
Artificial neural networks for learning robots
Sovka, Michal ; Jirků, Petr (advisor) ; Berka, Petr (referee)
The main aim of this bachelor thesis is to understand, describe and explain the basic principles and elements used in learning robots with using artificial neural networks. Firstly, I focus on robots in general and their basic functional units. Then it is to introduce the theory of learning applied to real environments. Nervous system in this work becomes a central learning muse about theory with using artificial neural networks. It's very important in understanding of artificial neurons and artificial neural networks as a complex I consider the biological neuron and its synapses. That's why I take them very seriously. After introduction to artificial neural networks I attend to only the one of their group fully used in robotics. Finally, I demonstrate the function of Kohonen type of artificial neural network used in robotics and scilicet in the application developed in one of a foreign university workplace.I hope the benefits of my work is in a comprehensive text focused on the basic elements of robotics, as well as artificial neural networks and their neurobiological assumption. The work can then be used in studying for people interested in a broader approach to intelligent robotics.
|
guest ::
